The overall objective of the proposed project is to rigorously develop novel methods for identifying, recovering and analyzing viable circulating tumor cells (CTCs). CTCs are the presumed precursors of the metastatic lesions responsible for the deaths of the vast majority of carcinoma patients, and the molecular and cellular analysis of CTCs could provide key insight into mechanisms underlying metastasis. While there are a variety of methods that are currently employed for identifying CTCs, most are limited because of their reliance on the ability to detect cell surface epithelial antigens. The loss of epithelial character with the epithelial-to-mesenchymal transition, as well as the ability of metastasis-promoting cell surface glycoproteins to interfere with antibody recognition of cell surface antigens, raises significant concerns as to the fidelity of these procedures. Here we will develop novel, more robust methods for identifying CTCs based on the altered metabolic state of cancer cells relative to non-transformed cells. The elevated ability of malignant cells to take up glucos and to extrude protons from their intracellular environment will be used as the basis for our assays to distinguish cancer from non-cancer cells. In Aim 1 we will assess the abilities of a fluorescent glucose derivative and an intracellular pH indicator to distinguish between malignant and non-transformed cells. These studies will be carried out with both adherent cells in culture and with malignant cells diluted into mouse blood. The studies of Aim 1 will establish that malignant tumor cells can be distinguished from non-transformed cells based on their metabolic properties, and will begin to develop the protocols necessary to identify CTCs in blood under experimental conditions. In Aim 2 we will apply the techniques developed in Aim 1 to a mouse model of breast cancer. In these studies we will determine whether CTCs derived from orthotopically implanted tumors can be identified using the metabolic sensors, and whether the newly developed methods are superior to conventional approaches. We anticipate that the successful completion of the proposed studies will provide investigators with the tools necessary to more reliably identify and recover CTCs from mouse models, and may ultimately prove beneficial in more precisely defining CTCs in the blood of carcinoma patients.